The Objective2 (O2) Headphone Amp DIY Project

[lenconian]

Originally Posted by regal

...but until you can answer the basic question whether an electron is a particle or a wave, I think we all need to keep an open mind and understand the limits of science.

Sorry for going Off Topic yet again here, but as you should know as an engineer, it's neither. An electron is a quantum object. Not a wave and not a particle. At times it can show behaviour, which we know as "wave"like or "particle"like in our macro world. But it can as well show both of these behaviours at the same time.

I'm sorry but I couldn't let that statement just stand there...

[regal]

Originally Posted by lenconian

Sorry for going Off Topic yet again here, but as you should know as an engineer, it's neither. An electron is a quantum object. Not a wave and not a particle. At times it can show behaviour, which we know as "wave"like or "particle"like in our macro world. But it can as well show both of these behaviours at the same time.

I'm sorry but I couldn't let that statement just stand there...

All I was pointing out is that science hasn't a unified theory, ie theory of relativity and quantum theory are mutually exclusive, one or both are wrong...

This being an "objectivist" amp I think we are right on topic. Objectivism takes a very a cartesian, newtonian world view. Works fine for calculating 99% of physical problems like falling apples to rocket trajectories, but we are dealing with hearing/human perception here, very challenging stuff.

I am all for using the dscope to help design an headphone amp, and am excited to hear it. But until science maps a working physical model of the ear's temporal membrane we must understand that any measurements of electronics where the goal is to reproduce accurately a live instrument/ensemble can't be a closed shut case of solved science. We would need a working model of perception.

I'm not taking sides and fully support the approach, just hate to see this 18th century mindset on a forum with so many well educated folks.

[lenconian]

I'm not taking sides and fully support the approach, just hate to see this 18th century mindset on a forum with so many well educated folks.

I don't think we're closed minded, why would you?
You know I agree with much of what you say in the grand picture. We're far from knowing it all and there's always room for speculation.
It's just the way you draw scientific references.
Do you really think there can and will be a perfect model which covers the human perception? With 6 billion people on the world how can that ever be?

You know we've come a long way with models of the human ear, just think about MP3 compression. And of course there will be further improvements on those models, but I'm a bit skeptical if we can get all that much better...

So all we can do is do the measurements we can, make sure our equipment has no major glitches and do the fine tuning by ear

Cheers.

[inarc]

Originally Posted by regal

But until science maps a working physical model of the ear's temporal membrane we must understand that any measurements of electronics where the goal is to reproduce accurately a live instrument/ensemble can't be a closed shut case of solved science.

The main objective of this amp is to reproduce a given signal (not instrument or ensemble) as close to the original as reasonable based on current scientific understanding. How a human perceives any such reproduction is peripheral to that, so your pseudophilosophical allusions to the problem of qualia are impertinent to begin with.

I had to register because I could no longer tolerate your trollish nonsense. Therefore, I ask you in the most polite tone possible right now to stop polluting this thread already and - if you are so inclined - take this discussion somewhere else where I can ignore it. I am interested in the OP's thread, not your personal drivel.

[iko]

If you will not refrain from personal comments points and bin time will be given. Final warning. All of you.

Your best bet is to stay on the exact topic this thread is about (title!).

[RocketScientist]

Originally Posted by regal

RocketScientist, I think you are confusing soundcard measurements with RMAA, soundcards with good software and a low noise buffered input can take very meaniful measurements..

The hardware of the dscope looks to me like an ADC with a low noise floor (what you find on $200+ soundcards). What's new is some brilliant software and buffers/interfaces to adjust power and input to the ADC, which I think is great but certainly not to the level of this holy war of objectivism, its basically a USB soundcard taken to a new level. What are they charging for a dScope?

Even if you want to talk about just soundcards, and ignore RMAA, there are still many problems and differences.

The dScope (or an Audio Precision analyzer) is far more than a "USB soundcard". Consider the "Continuous Time Analyzer" that allows for all sorts of real time measurements that can't be done with any sound card. Can a sound card extract the distortion residual signal and play it in real time over a speaker for example? Have a look at this Prism link.

Soundcards, and their drivers, have latency issues that prevent a lot of real time measurements. That's one reason Prism builds signal analysis into the dScope hardware. There's only so much you can do with a buffer full of non-real-time data in software. The dScope can also internally, and simultaneously, monitor its own outputs as well as inputs.

The dScope also has a whole range of digital capabilities soundcards do not. It can both analyze a digital signal for quality and flaws, and it can generate digital signals with controlled amounts of flaws. See here.

You gloss over the "interfaces" but the biggest problem is soundcards can't measure voltage. Even those calibrated in dB usually are not very accurate. So all you get with a soundcard are relative values which just doesn't work for many measurements. Sure you can use other instruments to set levels, but that only works until something changes, which can be often when making audio measurements. And you have to do lots of number crunching to get meaningful units for your results. It's a giant pain in the butt when the PC has no way to read absolute levels.

Can a sound card autorange the inputs with precision attenuators to always keep them within 2 dB of 0 dBFS while say doing a sweep? Soundcards have only crude attenuators or none at all. But it's critical when measuring something like the O2 with distortion so low it's only a few microvolts and flirting with the noise floor. Those same inputs that can measure down to a few microvolts can also accept nearly 200 volts. Automatically.

The above two paragraphs form a lethal combination. If you set soundcard levels using a DMM you can't change anything without needing to re-establish your references. But if you can't change anything, you can't keep the signal close to 0 dBFS to get the full dynamic range of the soundcard. And when trying to measure high performance gear, you need every dB of dynamic range you can get. This makes doing a THD vs output sweep of something like the O2 essentially impossible with a sound card.

The inputs and outputs are also fully isolated from chassis ground, USB ground, and each other. That's critical for measuring a lot of gear.

The inputs and outputs also have adjustable and well defined impedances which is critical for making accurate measurements. Soundcards don't.

The inputs also have adjustable hardware filtering for both low pass and high pass. Soundcards don't and you're stuck with their filtering which was never designed for making measurements. Aliasing is a real problem when you're trying to make some measurements but the anti-aliasing filters can corrupt other measurements.

Few soundcards allow external clocking/external sync which can be important for some digital testing.

There's more but my point is there are very significant hardware differences. I'm not saying a soundcard isn't useful for some measurements, but it's wishful thinking to believe a soundcard can do nearly everything a dScope or AP analyzer can if only someone had the right software.

[msjjr]

Hi everyone. I'm looking to join any future group buy, or if someone has a board unspoken for, I'd like to buy it. Thanks very much.

[slatedk]

Originally Posted by msjjr

Hi everyone. I'm looking to join any future group buy, or if someone has a board unspoken for, I'd like to buy it. Thanks very much.

Try here O2 Headphone Amplifier GB

Olli just handed in the order for 800 PCBs; there could be a few not spoken fore.

[msjjr]

Thank you. I just did.

[mauxigens]

@RocketScientist: I was just thinking... MrSlim wants to put more stuff on the front with the B3 version. He needs wires for this if I understand correctly. Could this degrade performance measurably?